The neurophysiological basis of visceral sensations in general and pain in particular have been mainly studied with short-lasting stimuli that simulate the acute events of visceral organ function. Pelvic viscera are supplied by spinal afferents which are involved in the coordinated reflex regulation of continence and evacuation of bowel and bladder and are capable of signalling impending or frank tissue damage. Typically, each afferent neuron innervates one viscus only. The organ-specific subtypes are functionally homogenous and encode by their discharge frequency the information for organ regulation, non-painful and painful sensations. Thus, pain elicited from these organs under physiological conditions is probably not elicited by a specific set of nociceptive visceral afferents. While the use of brief stimuli has yielded invaluable neurophysiological information for normal, healthy viscera, it has fallen somewhat short of providing information about the neuronal basis of chronic visceral pain states. Using pathological models such as experimental inflammation of the urinary bladder and ischaemia of the colon we have shown that the receptor properties of most afferents change dramatically. Of particular interest is the discovery of a novel type of visceral receptor which is not excited by extreme noxious mechanical stimuli applied to the healthy tissue but which is vigorously activated at the onset of an inflammation. This means that the number of functionally active primary afferents is not immutable, but critically depends on the state of the tissue. This new principle of plasticity in the peripheral nervous system bears some considerable importance for the understanding of the genesis of chronic visceral pain states.(ABSTRACT TRUNCATED AT 250 WORDS)